By the environmental DNA （eDNA） technique and physicochemical parameter analyses， the spatial distribution characteristics of biological community structure， habitat elements and pollutants in the surface sediments of Hangzhou Bay and their impacts on biodiversity were systematically analysed. A total of 53 phyla were identified， with Metazoa dominating （42.55%）， notably the key species Portunus trituberculatus （20.84%）， indicating strong keystone species-driven community assembly. Biodiversity exhibited distinct spatial gradients: species richness indices （Observed species: 513-3 169； Chao1: 777.33-4 185.46） decreased southward， while evenness （Pielou J: 0.05-0.79） and diversity （Shannon: 0.31-6.01） showed a north-high， central-low， and south-moderate pattern. The spatial differentiation of sediment grain size [mainly silt， （70.72±1.94）%] and TOC [（0.20±0.06）%] and TN [（0.06±0.01）%] showed that coarse-grained substrate drove community homogenisation by optimising crab habitat conditions， whereas C/N ratios （3.12±0.83） indicated the dominance of sea-sourced organic matter， and areas of high values of TOC and TN coincided with algal and protozoan dominance coincided.The analysis of contaminants showed that the contaminants in Hangzhou Bay sediments generally showed a decreasing offshore diffusion gradient， of which DDTs [（9.45±1.68） ng/g]， brominated flame retardants were dominated by NBFRs （76.7%）， BDE209 accounted for more than 71% of the PBDEs， and the PFCs and polyfluoroalkyl substances （PFCAs accounted for 87.9% of the PFCAs） were relatively homogeneously distributed in the sediment due to their high water solubility. The distribution of perfluorinated and polyfluorinated alkyl substances （including 87.9% of PFCAs） is relatively uniform in sediments due to high water solubility. The spatial heterogeneity of habitat elements and pollutants was shaped by tidal dynamics and land-based emissions. By integrating macro-scale redundancy analysis （RDA） and micro-ecological mechanism analysis， a synergistic pattern-mechanism verification system was constructed in Hangzhou Bay. The macro-scale RDA reveals the response of biological community diversity to multi-dimensional environmental gradients， while the micro-scale RDA illustrates the differential driving effects of pollutants and environmental factors on functional communities. The mutual validation of the two systems shows that the cross-scale mutual feedback mechanism is the core pathway to analyse the emergent characteristics of ecosystems， and provides a scientific basis for the multi-stress synergistic management of large-scale strong tidal estuarine ecosystems.